Racial implications of time to surgery in disparities in thyroid cancer survival

INTRODUCTION

The influence of time to surgery on racial/ethnic disparities in papillary thyroid carcinoma (PTC) survival remains unstudied.

MATERIALS AND METHODS

The National Cancer Database (2004-2017) was queried for patients with localized PTC. Survival data was compared by time to surgery, patient demographics, and multivariable Cox regression was performed.

RESULTS

Of 126,708 patients included, 5% were Black, 10% Hispanic. Of all patients, 85% had no comorbidities. Non-Hispanic White (NHW) patients had a shorter median time to surgery than Black and Hispanic patients (36 vs. 43 vs. 42 days, respectively p ​< ​0.001). In multivariable analysis, longer time to surgery (>90 days vs ​< ​30 days) and Black race vs NHW, were associated with worse survival (HR: 1.56, (95%CI, 1.43-1.70), p ​< ​0.001 and HR: 1.21, (1.08-1.36), p ​= ​0.001), respectively.

CONCLUSION

Delaying surgery for thyroid cancer is associated with worse survival. However, independent of time to surgery and other confounders, there remains a disparity as black patients have poorer outcomes.

Hub and spoke framework for study of surgical centralization within United States health systems

BACKGROUND

Hospital consolidation into health systems has mixed effects on surgical quality, potentially related to degree of surgical centralization at high-volume (hub) sites. We developed a novel measure of centralization and evaluated a hub and spoke framework.

METHODS

Surgical centralization within health systems was measured using hospital surgical volumes (American Hospital Association) and health system data (Agency for Healthcare Research and Quality). Hub and spoke hospitals were compared using mixed effects logistic regression and system characteristics associated with surgical centralization were identified using a linear model.

RESULTS

Within 382 health systems containing 3022 hospitals, system hubs perform 63% of cases (IQR 40-84%). Hubs are larger, in metropolitan and urban areas, and more often academically affiliated. Degree of surgical centralization varies ten-fold. Larger, multistate, and investor-owned systems are less centralized. Adjusting for these factors, there is less centralization among teaching systems (p ​< ​0.001).

CONCLUSIONS

A hub-spoke framework applies to most health systems but centralization varies significantly. Future studies of health system surgical care should assess the contributions of surgical centralization and teaching status on differential quality.

Production, purification, characterization, and applications of -galactosidase from Bacillus flexus JS27 isolated from Manikaran hot springs

α-Galactosidase hydrolyzes the α-1,6-linkage present at the non-reducing end of the sugars and results in the release of galactosyl residue from oligosaccharides like melibiose, raffinose, stachyose, etc. In the present study we report, α-galactosidase from Bacillus flexus isolated from Manikaran hot springs (India). Maximum enzyme production was obtained in guar gum and soybean meal after 72 h at 150 rpm. While, the temperature/pH of production was optimized at 50 °C and 7.0, respectively. Isoenzymes (α-gal I and II) were obtained and characterized based on temperature/pH optima along with their stability profile. JS27 α-Gal II was purified with a final purification fold of 11.54. Native and SDS-PAGE were used to determine the molecular weight of the enzyme as 86 and 41 kDa, respectively, indicating its homodimeric form. JS27 α-Gal II showed optimum enzyme activity at 55 °C and pH 7 (10 min). The enzyme displayed K_m value of 2.3809 mM and V_max of 2.0 × 10⁴ µmol/min/ml with pNPG as substrate. JS27 α-Gal II demonstrated substrate hydrolysis and simultaneous formation of transgalactosylation products (α-GOS) with numerous substrates (sugar/sugar alcohols, oligosaccharides, and complex carbohydrates) which were verified by TLC and HPLC analysis. α-GOS are significant functional food ingredients and can be explored as prebiotics.

Evaluation of prognostic risk models for postoperative pulmonary complications in adult patients undergoing major abdominal surgery: a systematic review and international external validation cohort study

BACKGROUND

Stratifying risk of postoperative pulmonary complications after major abdominal surgery allows clinicians to modify risk through targeted interventions and enhanced monitoring. In this study, we aimed to identify and validate prognostic models against a new consensus definition of postoperative pulmonary complications.

METHODS

We did a systematic review and international external validation cohort study. The systematic review was done in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We searched MEDLINE and Embase on March 1, 2020, for articles published in English that reported on risk prediction models for postoperative pulmonary complications following abdominal surgery. External validation of existing models was done within a prospective international cohort study of adult patients (≥18 years) undergoing major abdominal surgery. Data were collected between Jan 1, 2019, and April 30, 2019, in the UK, Ireland, and Australia. Discriminative ability and prognostic accuracy summary statistics were compared between models for the 30-day postoperative pulmonary complication rate as defined by the Standardised Endpoints in Perioperative Medicine Core Outcome Measures in Perioperative and Anaesthetic Care (StEP-COMPAC). Model performance was compared using the area under the receiver operating characteristic curve (AUROCC).

FINDINGS

In total, we identified 2903 records from our literature search; of which, 2514 (86·6%) unique records were screened, 121 (4·8%) of 2514 full texts were assessed for eligibility, and 29 unique prognostic models were identified. Nine (31·0%) of 29 models had score development reported only, 19 (65·5%) had undergone internal validation, and only four (13·8%) had been externally validated. Data to validate six eligible models were collected in the international external validation cohort study. Data from 11 591 patients were available, with an overall postoperative pulmonary complication rate of 7·8% (n=903). None of the six models showed good discrimination (defined as AUROCC ≥0·70) for identifying postoperative pulmonary complications, with the Assess Respiratory Risk in Surgical Patients in Catalonia score showing the best discrimination (AUROCC 0·700 [95% CI 0·683-0·717]).

INTERPRETATION

In the pre-COVID-19 pandemic data, variability in the risk of pulmonary complications (StEP-COMPAC definition) following major abdominal surgery was poorly described by existing prognostication tools. To improve surgical safety during the COVID-19 pandemic recovery and beyond, novel risk stratification tools are required.

FUNDING

British Journal of Surgery Society.

AB0307 MERKEL CELL CARCINOMA RECURRENCE RISK IS LOWER IN PATIENTS WITH AUTOIMMUNE DISEASE THAN IN THOSE WITH OTHER TYPES OF IMMUNE SUPPRESSION

Background

Merkel cell carcinoma (MCC) is twice as likely to recur in immunosuppressed (IS) patients as in immunocompetent (IC) patients. Iatrogenic IS due to autoimmune disease (AD) may influence prognosis differently than intrinsic IS such as due to hematologic malignancy. Moreover, modification of IS medication may improve prognosis.

Objectives

Our objective was to evaluate the risk of MCC recurrence among patients with AD diseases relative to other immunosuppressive conditions among 762 MCC patients from an Institutional review board-approved registry.

Methods

We categorized patients into 3 groups: IS due to AD (ISAD); IS from other causes (lSnon-AD) such as hematologic malignancy, solid organ transplant, human immunodeficiency virus; or immune competent (n=31, 70, and 661 respectively). ISAD patients were subcategorized into rheumatoid arthritis (RA) (ADRA, n=13) vs. AD except for RA (ADnon-RA, n=18). Descriptive statistics were used to compare the features of different characteristics in each group. Kaplan-Meier survival curves were constructed to assess the cumulative incidence of recurrence in different patient groups. In order to estimate the associations between baseline patient characteristics and the risk of MCC recurrence, Fine and Gray regression models were used with death as a competing risk for recurrence. The multivariable models adjusted for age, sex, and extent of MCC at initial presentation.

Results

Patients with ISAD had lower stage disease (local disease: 58% vs. 36%, p = 0.003) and smaller primary tumors than ISnon-AD (<= 2 cm: 83% vs. 57%, p=0.023). After adjusting for age, sex, and stage, ISAD patients (ADRA and ADnon-RA) overall had a 54% higher recurrence rate (hazard ratio (HR): 1.54, p=0.21) than IC patients. In comparison, ISnon-AD group had a 165% higher recurrence rate (HR: 2.65, p<0.001) than IC patients (Figure 1). When considered separately, ADRA pts appeared to have a similar recurrence rate as IC pts (HR: 1.19, p=0.76) while ADnon-RA pts had a higher recurrence rate (HR: 1.83, p=0.16) relative to IC pts. At the time of MCC diagnosis, 80% (n=24) of AD pts were on IS medication including conventional disease modifying drugs, biologics, or oral steroids. After MCC diagnosis, 22% (5 patients) stopped all immunosuppressive medications. Among patients on biologics, 89% (8/9 pts) elected to stop the drug. Eleven pts with AD experienced recurrences. Our study was underpowered to demonstrate associations regarding use of a particular immunosuppressive medication and MCC recurrence.

Figure 1.

Cumulative incidence of Merkel cell carcinoma recurrence in different patient groups.

AD = autoimmune disease; IS = immunosuppressed; RA = rheumatoid arthritis.There were 4/13, 8/18, 49/70, and 217/661 recurrences in the RA, ADnon-RA, other immunosuppressed, and not chronically immunosuppressed groups, respectively. Follow-up time ranged from 26 days to 16 years, with median follow-up times of 4.7 years, 1.6 years, 1.6 years, and 3.9 years for the RA, ADnon-RA, other immunosuppressed and not chronically immunosuppressed groups, respectively.

Conclusion

In this cohort, pts with AD appeared to have a better prognosis than intrinsic IS, with RA conferring very little risk above that for immune competent pts.

Acknowledgements

I have no acknowledgements to declare.

Disclosure of Interests

None declared

Insights on sustainable approaches for production and applications of value added products

The increasing demand for the development of sustainable strategies to utilize and process agro-industrial residues paves new paths for exploring innovative approaches in this area. Biotechnology based microbial transformations provide efficient, low cost and sustainable approaches for the production of value added products. The use of organic rich residues opens new avenues for the production of enzymes, pigments, biofuels, bioactive compounds, biopolymers etc. with vast industrial and therapeutic applications. Innovative technologies like strain improvement, enzyme immobilization, genome editing, morphological engineering, ultrasound/supercritical fluid/pulse electric field extraction, etc. can be employed. These will be helpful in achieving significant improvement in qualitative and quantitative parameters of the finished products. The global trend for the valorisation of biowaste has boosted the commercialization of these products which has transformed the markets by providing new investment opportunities. The upstream processing of raw materials using microbes poses a limitation in terms of product development and recovery which can be overcome by modifying the bioreactor design, physiological parameters or employing alternate technologies which will be discussed in this review. The other problems related to the processes include product stability, industrial applicability and cost competitiveness which needs to be addressed. This review comprehensively discusses the recent progress, avenues and challenges in the approaches aimed at valorisation of agro-industrial wastes along with possible opportunities in the bioeconomy.

Identification of other Cellular Receptors for Edema Factor of Bacillus anthracis by Independent Inhibition of Protective Antigen Evidenced by Inhibition of Embryo Growth and Angiogenesis

Edema factor (EF) is one of the major secretory proteins of anthrax bacteria along with protective antigen (PA) and lethal factor (LF). Edema factor is a calmodulin-and calcium-dependent adenylate cyclase that increases intracellular levels of cAMP. Intracellular trafficking of EF occurs through PA by binding to ATR/CMG2 receptors, which are also involved in other physiological functions of cells. cAMP is a secondary messenger which activates multiple signaling cascades involved in the cytokinetics of actin molecules and cell junction formation. The present study evaluated the effect of EF on growth and angiogenesis patterns in chicken embryos in the in ovo model. Angiogenesis in the chorioallantoic membrane (CAM) of an embryonated chicken egg was decreased and embryo growth was delayed by EF despite the absence of trafficking moiety PA, which is required for transferring the EF molecule inside the cell. Angiogenesis inhibition and embryo growth retardation indicate the use of an alternative receptor by EF to modulate these cellular functions. Additionally, docking was performed between EF as a ligand and hepatocyte growth factor receptor (cMET) and vascular endothelial growth factor (VEGF) receptors, which are mainly involved in growth and angiogenesis. The analysis revealed a very strong binding of EF to cMET receptor (in terms of the number of hydrogen bonds and energy) compared to its ligand hepatocyte growth factor (HGF), which indicates the use of cMET receptor by EF and induction of angiogenesis and embryo growth retardation possibly by competitive inhibition of HGF ligand or receptor-mediated endocytosis.

Avelumab in patients with previously treated metastatic Merkel cell carcinoma (JAVELIN Merkel 200): updated overall survival data after >5 years of follow-up

BACKGROUND

Merkel cell carcinoma (MCC) is a rare, aggressive skin cancer that has a poor prognosis in patients with advanced disease. Avelumab [anti-programmed death-ligand 1 (PD-L1)] became the first approved treatment for patients with metastatic MCC (mMCC), based on efficacy and safety data observed in the JAVELIN Merkel 200 trial. We report long-term overall survival (OS) data after >5 years of follow-up from the cohort of patients with mMCC whose disease had progressed after one or more prior lines of chemotherapy.

PATIENTS AND METHODS

In Part A of the single-arm, open-label, phase II JAVELIN Merkel 200 trial, patients with mMCC that had progressed following one or more prior lines of chemotherapy received avelumab 10 mg/kg by intravenous infusion every 2 weeks until confirmed disease progression, unacceptable toxicity, or withdrawal. In this analysis, long-term OS was analyzed.

RESULTS

In total, 88 patients were treated with avelumab. At data cut-off (25 September 2020), median follow-up was 65.1 months (range 60.8-74.1 months). One patient (1.1%) remained on treatment, and an additional patient (1.1%) had reinitiated avelumab after previously discontinuing treatment. Median OS was 12.6 months [95% confidence interval (CI) 7.5-17.1 months], with a 5-year OS rate of 26% (95% CI 17% to 36%). In patients with PD-L1+ versus PD-L1- tumors, median OS was 12.9 months (95% CI 8.7-29.6 months) versus 7.3 months (95% CI 3.4-14.0 months), and the 5-year OS rate was 28% (95% CI 17% to 40%) versus 19% (95% CI 5% to 40%), respectively (HR 0.67; 95% CI 0.36-1.25).

CONCLUSION

Avelumab monotherapy resulted in meaningful long-term OS in patients with mMCC whose disease had progressed following chemotherapy. These results further support the role of avelumab as a standard of care for patients with mMCC.

1275 Machine Learning for Outcome Prediction in Epilepsy Surgery: A Systematic Review

Aim

A third of epilepsy patients suffer from medically refractory seizures. In patients eligible for surgical treatment, seizure freedom rates remain variable. Machine learning (ML) utilises large datasets to detect patterns to make predictions. We systematically review studies employing ML models for prediction of outcome following resective epilepsy surgery to evaluate their efficacy, applicability and value in determining surgical candidacy.

Method

MEDLINE, Cochrane and EMBASE databases were searched for literature published between 2010 – 2020 according to PRISMA guidance. Non-refractory epilepsy, non-clinical outcome prediction, or non-human studies were excluded. Clinical and demographic data, ML features, discrimination and prediction accuracy metrics were extracted.

Results

15 studies were included. Median cohort size was 49 (range 16 – 4211). Heterogeneous input data sources were utilised: MRI (n = 10) , electrophysiology (n = 4), PET (n = 2), clinical data (n = 2), and neuropsychological testing (n = 1). The most common ML model used was support vector machines (n = 7). All studies had good discrimination (AUC &gt; 0.70, range: 0.79 [95% CI NR] - 0.94 [95% CI 0.92 – 0.96]), and good prediction accuracy (&gt; 0.70, range: 0.76 [95% CI NR] – 0.95 [95% CI NR]). Limitations included small sample sizes, limited external validation and lack of comparison with clinician-predicted outcomes.

Conclusions

Machine Learning for outcome prediction could enhance clinical decision-making for surgical candidacy in epilepsy, and lead to improved precision medicine delivery. Outcome reporting remains inconsistent, and further work is required to externally validate such models to implement these to large-scale clinical populations.

A pre-registered short-term forecasting study of COVID-19 in Germany and Poland during the second wave

Disease modelling has had considerable policy impact during the ongoing COVID-19 pandemic, and it is increasingly acknowledged that combining multiple models can improve the reliability of outputs. Here we report insights from ten weeks of collaborative short-term forecasting of COVID-19 in Germany and Poland (12 October-19 December 2020). The study period covers the onset of the second wave in both countries, with tightening non-pharmaceutical interventions (NPIs) and subsequently a decay (Poland) or plateau and renewed increase (Germany) in reported cases. Thirteen independent teams provided probabilistic real-time forecasts of COVID-19 cases and deaths. These were reported for lead times of one to four weeks, with evaluation focused on one- and two-week horizons, which are less affected by changing NPIs. Heterogeneity between forecasts was considerable both in terms of point predictions and forecast spread. Ensemble forecasts showed good relative performance, in particular in terms of coverage, but did not clearly dominate single-model predictions. The study was preregistered and will be followed up in future phases of the pandemic.

Odderon Exchange from Elastic Scattering Differences between pp and pp[over ¯] Data at 1.96 TeV and from pp Forward Scattering Measurements

We describe an analysis comparing the pp[over ¯] elastic cross section as measured by the D0 Collaboration at a center-of-mass energy of 1.96 TeV to that in pp collisions as measured by the TOTEM Collaboration at 2.76, 7, 8, and 13 TeV using a model-independent approach. The TOTEM cross sections, extrapolated to a center-of-mass energy of sqrt[s]=1.96  TeV, are compared with the D0 measurement in the region of the diffractive minimum and the second maximum of the pp cross section. The two data sets disagree at the 3.4σ level and thus provide evidence for the t-channel exchange of a colorless, C-odd gluonic compound, also known as the odderon. We combine these results with a TOTEM analysis of the same C-odd exchange based on the total cross section and the ratio of the real to imaginary parts of the forward elastic strong interaction scattering amplitude in pp scattering for which the significance is between 3.4σ and 4.6σ. The combined significance is larger than 5σ and is interpreted as the first observation of the exchange of a colorless, C-odd gluonic compound.

875 Use of Oriented Priors Through Magnetic Tractography (MAGNET) In Deliniation Of Meyer's Loop and Correlation with Visual Field Deficit In Temporal Lobe Epilepsy (TLE) Surgery: A Pilot Study

Introduction

Pre-operative white matter tract reconstruction of the Meyer’s loop (ML) of the optic radiation using diffusion MRI (tractography) can be used to prevent post-operative visual-field deficit. Due to its complex anatomy, precise reconstruction of the ML is challenging and often underestimated. Previous work has suggested that an innovative tractography technique using oriented priors called MAGNET better approximates reconstruction to reported histological prosections. This proof-of-context study validates the MAGNET methodology in predicting visual-field deficits in patients undergoing TLE surgery.

Method

Diffusion MRI datasets were used to reconstruct pre-operative ML using MAGNET in five patients. These were overlaid on post-operative T2-MRI series demonstrating the surgically resected area to measure overlap between resection and reconstructed ML. A correlation with post-operative visual-field defects was established.

Results

There was no evidence of visual field deficit in the cases where there was no overlap between the reconstructed ML and the resected region. In the cases with overlap with reconstructed ML and resection, there was visual deficit found. There was no correlation between proportion of resected ML and visual deficit.

Conclusions

This pilot demonstrates that MAGNET accurately reconstructs ML in pre-surgical TLE cases compared to standard tractography techniques and can be used to augment neurosurgical planning and resection.

Pathological and immunological protection induced by inactivated reverse genetics-based H3N8 equine influenza vaccine candidate in murine model

Equine influenza (EI) is an important viral respiratory disease of equines caused by influenza A virus (IAV). The antigenic drift in IAVs necessitates regular updating and harmonization of vaccine strain with the circulating virus. The reverse genetics-based recombinant viruses could be easy instrument in generating vaccine against circulating virus in a quick and effective manner. Present study has been envisaged to evaluate the immunogenicity and protective efficacy of inactivated recombinant equine influenza virus (rgEIV) vaccine candidate having six segments from H1N1 virus (A/WSN/33/H1N1) and HA (hemaglutinin) and NA (neuraminidase) segments from H3N8 equine influenza virus [(A/eq/Jammu-Katra/06/08) of clade 2 of Florida sublineage] generated through reverse genetic engineering. BALB/c mice were immunized with inactivated rgEIV adjuvanted with aluminium hydroxide gel and challenged with H3N8 virus (A/eq/Jammu-Katra/06/08). The protective efficacy was evaluated through serology, cytokine profiling, clinical signs, gross and histopathological changes, immunohistochemistry and residual virus quantification. Immunizations induced robust humoral immune response as estimated through hemagglutination inhibition assay (HAI). The antibodies were isotyped and the predominant subclass was IgG1. The vaccine candidate produced mixed Th1 and Th2 responses through stimulation of IFN-γ, IL-2, IL-4 and IL-6 expression. Immunization protected mice against challenge as reflected through reduction in clinical signs and body weight loss, early recovery, mild pathological changes (gross and histopathological lesions) as evident through scoring of lesions, low residual virus in nasopharynx and lungs quantified through egg titration and quantitative reverse transcriptase PCR (qRT-PCR). The study demonstrates that inactivated recombinant EIV generated through reverse genetic approach provides equivalent protection to that observed with inactivated whole H3N8 EIV vaccine. Keywords: equine influenza; reverse genetics; vaccine; pathology; murine model.

Introduction of Laparoscopic Donor Nephrectomy: Challenges, Outcomes and Success Strategies

BACKGROUND

Laparoscopic live donor nephrectomy (LLDN) has become the standard of care and is popular among most of the transplant centers across the globe. Despite proven advantages of LLDN, some transplantation centers hesitate to start the program because of issues concerning donor safety and allograft function.

OBJECTIVE

To discusses the main barriers for creating a successful LLDN program, strategies that allowed us to start a successful LLDN program along with the study results.

METHODS

The donors undergoing LLDN from December 2016 to February 2018 were enrolled in the study and prospectively evaluated. LLDN were performed by two senior surgeons alternately with assistance by the laparoscopic urologist in all cases. Also, in the present study, two technical alterations were done in the standard surgical technique of transperitoneal LDN. The first important modification made was the use of two additional ports for use by laparoscopic urologists. The second modification involved dissection on both poles of the kidney before hilar dissection.

RESULTS

A total of 112 transperitoneal LLDN were performed during the study period. The mean (range) of operation time was 117.5 (81-158) min; the ischemia time was 194 (171-553) sec. Only one patient needed conversion to open surgery. No other major peri-operative or posto-perative complications occurred. All kidney grafts were functioning well.

CONCLUSION

With proper planning, team approach, and few technical modifications, introduction of LLDN is safe and effective.

Cocaine use and HIV-infection modify coronary plaque morphology differently than conventional cardiovascular risk factors

Background/Introduction

Cardiovascular risk factors are assumed to propagate coronary artery disease (CAD) through the lipid and/or inflammatory pathologic pathways. However, we currently do not have any in-vivo information whether these effects may alter CAD morphology and structure differently.

Purpose

We wished to assess whether conventional cardiovascular risk factors (which are assumed to affect CAD primarily through the lipid pathway), and unconventional risk factors such as cocaine use and human immunodeficiency virus (HIV)-infection (which are assumed to propagate CAD primarily through the inflammatory pathway) affect different coronary plaque structures as assessed by radiomics.

Methods

In our prospective longitudinal observational study of 300 individuals without cardiovascular symptoms but with coronary CT angiography (CCTA)-confirmed atherosclerosis (210 male, age: 48±7 years) of whom 161 were cocaine users at baseline with or without HIV-infection (226 HIV-infected), underwent CCTA at two time points (mean follow-up: 4.0±2.3 years). Precision phenotyping of CAD was done by calculating 1276 radiomic features on the 861 plaques. Linear mixed models corrected for plaque volume, high-sensitivity C-reactive protein, statin use and positive family history were used to assess the effects of chronic cocaine use, HIV-infection and elevated atherosclerotic cardiovascular disease risk (ASCVD≥7.5%). Hierarchical clustering was used to assess potential clusters among significant radiomic features. Bonferroni corrected p&lt;0.00004 (0.05/1276) was considered significant.

Results

Overall, 32.0% (409/1276) of the radiomic features showed any significant association, of which 74.1% (303/409), 4.2% (17/409) and 25.4% (104/409) were affected by cocaine use, HIV-infection and elevated ASCVD risk, respectively. There was no overlap among radiomic features significantly associated with increased ASCVD risk and cocaine use or HIV-infection, while 88,2% (15/17) of HIV-infection associated parameters were also affected by cocaine use. Cluster analysis indicated 13 different structural components among significant features, of which eight were unique to chronic cocaine use, three unique to ASCVD risk, and two contained parameters associated with chronic cocaine use, elevated ASCVD risk and/or HIV-infection.

Interpretation

Chronic cocaine use and HIV-infection modify different CAD morphological components than conventional cardiovascular risk factors, potentially implying independent pathological pathways of disease progression irrespective of each other.

Funding Acknowledgement

Type of funding source: Public grant(s) – National budget only. Main funding source(s): National Institutes of Health, National Institute on Drug Abuse